The availability of various content items on the Internet allows access to information in bulk. However, the sheer volume of content items available does not increase the usefulness of the content items. The information presented in the content items themselves must be suitable for each particular user. For example, providing content items with information on living and style may be irrelevant to a user interested in learning about current political developments, even though the same content items may be enjoyable for another user.
However, the fragmented nature of content from many different content item providers, as well as the number of exchanges where requests may originate, complicates the handling of requests. For example, it is difficult for a request processing system to consolidate all the content items from a variety of content item providers to respond to requests and understand whether outcomes to the requests were successful. Further, because there may be one or more exchanges where requests originate, it is difficult to scale the request processing system to respond to the sheer quantity of requests from more than one exchange at a time. Oftentimes, the request processing system becomes unable to accommodate every request, as the volume of requests may exceed the throughput of the request processing system. This problem is compounded when the request processing system would like to maintain a minimum level of service to each exchange, as overall service capacity may be severely exhausted.
Therefore, it is desirable to improve the throughput of request processing systems, while allowing understanding of the outcomes of responses to requests.
The approaches described in this section are approaches that could be pursued, but not necessarily approaches that have been previously conceived or pursued. Therefore, unless otherwise indicated, it should not be assumed that any of the approaches described in this section qualify as prior art merely by virtue of their inclusion in this section.